[linux-2.6-block.git] / fs / udf / unicode.c

/*
 * unicode.c
 *
 * PURPOSE
 *	Routines for converting between UTF-8 and OSTA Compressed Unicode.
 *      Also handles filename mangling
 *
 * DESCRIPTION
 *	OSTA Compressed Unicode is explained in the OSTA UDF specification.
 *		http://www.osta.org/
 *	UTF-8 is explained in the IETF RFC XXXX.
 *		ftp://ftp.internic.net/rfc/rfcxxxx.txt
 *
 * COPYRIGHT
 *	This file is distributed under the terms of the GNU General Public
 *	License (GPL). Copies of the GPL can be obtained from:
 *		ftp://prep.ai.mit.edu/pub/gnu/GPL
 *	Each contributing author retains all rights to their own work.
 */

#include "udfdecl.h"

#include <linux/kernel.h>
#include <linux/string.h>	/* for memset */
#include <linux/nls.h>
#include <linux/crc-itu-t.h>

#include "udf_sb.h"

static int udf_translate_to_linux(uint8_t *, uint8_t *, int, uint8_t *, int);

static int udf_char_to_ustr(struct ustr *dest, const uint8_t *src, int strlen)
{
	if ((!dest) || (!src) || (!strlen) || (strlen > UDF_NAME_LEN - 2))
		return 0;

	memset(dest, 0, sizeof(struct ustr));
	memcpy(dest->u_name, src, strlen);
	dest->u_cmpID = 0x08;
	dest->u_len = strlen;

	return strlen;
}

/*
 * udf_build_ustr
 */
int udf_build_ustr(struct ustr *dest, dstring *ptr, int size)
{
	int usesize;

	if (!dest || !ptr || !size)
		return -1;
	BUG_ON(size < 2);

	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
	usesize = min(usesize, size - 2);
	dest->u_cmpID = ptr[0];
	dest->u_len = usesize;
	memcpy(dest->u_name, ptr + 1, usesize);
	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);

	return 0;
}

/*
 * udf_build_ustr_exact
 */
static int udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
{
	if ((!dest) || (!ptr) || (!exactsize))
		return -1;

	memset(dest, 0, sizeof(struct ustr));
	dest->u_cmpID = ptr[0];
	dest->u_len = exactsize - 1;
	memcpy(dest->u_name, ptr + 1, exactsize - 1);

	return 0;
}

/*
 * udf_ocu_to_utf8
 *
 * PURPOSE
 *	Convert OSTA Compressed Unicode to the UTF-8 equivalent.
 *
 * PRE-CONDITIONS
 *	utf			Pointer to UTF-8 output buffer.
 *	ocu			Pointer to OSTA Compressed Unicode input buffer
 *				of size UDF_NAME_LEN bytes.
 * 				both of type "struct ustr *"
 *
 * POST-CONDITIONS
 *	<return>		Zero on success.
 *
 * HISTORY
 *	November 12, 1997 - Andrew E. Mileski
 *	Written, tested, and released.
 */
int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
{
	const uint8_t *ocu;
	uint8_t cmp_id, ocu_len;
	int i;

	ocu_len = ocu_i->u_len;
	if (ocu_len == 0) {
		memset(utf_o, 0, sizeof(struct ustr));
		return 0;
	}

	cmp_id = ocu_i->u_cmpID;
	if (cmp_id != 8 && cmp_id != 16) {
		memset(utf_o, 0, sizeof(struct ustr));
		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
		       cmp_id, ocu_i->u_name);
		return 0;
	}

	ocu = ocu_i->u_name;
	utf_o->u_len = 0;
	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {

		/* Expand OSTA compressed Unicode to Unicode */
		uint32_t c = ocu[i++];
		if (cmp_id == 16)
			c = (c << 8) | ocu[i++];

		/* Compress Unicode to UTF-8 */
		if (c < 0x80U)
			utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
		else if (c < 0x800U) {
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0xc0 | (c >> 6));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 | (c & 0x3f));
		} else {
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0xe0 | (c >> 12));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 |
							  ((c >> 6) & 0x3f));
			utf_o->u_name[utf_o->u_len++] =
						(uint8_t)(0x80 | (c & 0x3f));
		}
	}
	utf_o->u_cmpID = 8;

	return utf_o->u_len;
}

/*
 *
 * udf_utf8_to_ocu
 *
 * PURPOSE
 *	Convert UTF-8 to the OSTA Compressed Unicode equivalent.
 *
 * DESCRIPTION
 *	This routine is only called by udf_lookup().
 *
 * PRE-CONDITIONS
 *	ocu			Pointer to OSTA Compressed Unicode output
 *				buffer of size UDF_NAME_LEN bytes.
 *	utf			Pointer to UTF-8 input buffer.
 *	utf_len			Length of UTF-8 input buffer in bytes.
 *
 * POST-CONDITIONS
 *	<return>		Zero on success.
 *
 * HISTORY
 *	November 12, 1997 - Andrew E. Mileski
 *	Written, tested, and released.
 */
static int udf_UTF8toCS0(dstring *ocu, struct ustr *utf, int length)
{
	unsigned c, i, max_val, utf_char;
	int utf_cnt, u_len;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;

try_again:
	u_len = 0U;
	utf_char = 0U;
	utf_cnt = 0U;
	for (i = 0U; i < utf->u_len; i++) {
		c = (uint8_t)utf->u_name[i];

		/* Complete a multi-byte UTF-8 character */
		if (utf_cnt) {
			utf_char = (utf_char << 6) | (c & 0x3fU);
			if (--utf_cnt)
				continue;
		} else {
			/* Check for a multi-byte UTF-8 character */
			if (c & 0x80U) {
				/* Start a multi-byte UTF-8 character */
				if ((c & 0xe0U) == 0xc0U) {
					utf_char = c & 0x1fU;
					utf_cnt = 1;
				} else if ((c & 0xf0U) == 0xe0U) {
					utf_char = c & 0x0fU;
					utf_cnt = 2;
				} else if ((c & 0xf8U) == 0xf0U) {
					utf_char = c & 0x07U;
					utf_cnt = 3;
				} else if ((c & 0xfcU) == 0xf8U) {
					utf_char = c & 0x03U;
					utf_cnt = 4;
				} else if ((c & 0xfeU) == 0xfcU) {
					utf_char = c & 0x01U;
					utf_cnt = 5;
				} else {
					goto error_out;
				}
				continue;
			} else {
				/* Single byte UTF-8 character (most common) */
				utf_char = c;
			}
		}

		/* Choose no compression if necessary */
		if (utf_char > max_val) {
			if (max_val == 0xffU) {
				max_val = 0xffffU;
				ocu[0] = (uint8_t)0x10U;
				goto try_again;
			}
			goto error_out;
		}

		if (max_val == 0xffffU)
			ocu[++u_len] = (uint8_t)(utf_char >> 8);
		ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
	}

	if (utf_cnt) {
error_out:
		ocu[++u_len] = '?';
		printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	}

	ocu[length - 1] = (uint8_t)u_len + 1;

	return u_len + 1;
}

static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o,
			const struct ustr *ocu_i)
{
	const uint8_t *ocu;
	uint8_t cmp_id, ocu_len;
	int i, len;


	ocu_len = ocu_i->u_len;
	if (ocu_len == 0) {
		memset(utf_o, 0, sizeof(struct ustr));
		return 0;
	}

	cmp_id = ocu_i->u_cmpID;
	if (cmp_id != 8 && cmp_id != 16) {
		memset(utf_o, 0, sizeof(struct ustr));
		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
		       cmp_id, ocu_i->u_name);
		return 0;
	}

	ocu = ocu_i->u_name;
	utf_o->u_len = 0;
	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
		/* Expand OSTA compressed Unicode to Unicode */
		uint32_t c = ocu[i++];
		if (cmp_id == 16)
			c = (c << 8) | ocu[i++];

		len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
				    UDF_NAME_LEN - utf_o->u_len);
		/* Valid character? */
		if (len >= 0)
			utf_o->u_len += len;
		else
			utf_o->u_name[utf_o->u_len++] = '?';
	}
	utf_o->u_cmpID = 8;

	return utf_o->u_len;
}

static int udf_NLStoCS0(struct nls_table *nls, dstring *ocu, struct ustr *uni,
			int length)
{
	int len;
	unsigned i, max_val;
	uint16_t uni_char;
	int u_len;

	memset(ocu, 0, sizeof(dstring) * length);
	ocu[0] = 8;
	max_val = 0xffU;

try_again:
	u_len = 0U;
	for (i = 0U; i < uni->u_len; i++) {
		len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
		if (!len)
			continue;
		/* Invalid character, deal with it */
		if (len < 0) {
			len = 1;
			uni_char = '?';
		}

		if (uni_char > max_val) {
			max_val = 0xffffU;
			ocu[0] = (uint8_t)0x10U;
			goto try_again;
		}

		if (max_val == 0xffffU)
			ocu[++u_len] = (uint8_t)(uni_char >> 8);
		ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
		i += len - 1;
	}

	ocu[length - 1] = (uint8_t)u_len + 1;
	return u_len + 1;
}

int udf_get_filename(struct super_block *sb, uint8_t *sname, uint8_t *dname,
		     int flen)
{
	struct ustr *filename, *unifilename;
	int len = 0;

	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	if (!filename)
		return 0;

	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	if (!unifilename)
		goto out1;

	if (udf_build_ustr_exact(unifilename, sname, flen))
		goto out2;

	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		if (!udf_CS0toUTF8(filename, unifilename)) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			goto out2;
		}
	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
		if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
				  unifilename)) {
			udf_debug("Failed in udf_get_filename: sname = %s\n",
				  sname);
			goto out2;
		}
	} else
		goto out2;

	len = udf_translate_to_linux(dname, filename->u_name, filename->u_len,
				     unifilename->u_name, unifilename->u_len);
out2:
	kfree(unifilename);
out1:
	kfree(filename);
	return len;
}

int udf_put_filename(struct super_block *sb, const uint8_t *sname,
		     uint8_t *dname, int flen)
{
	struct ustr unifilename;
	int namelen;

	if (!udf_char_to_ustr(&unifilename, sname, flen))
		return 0;

	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
		namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
		if (!namelen)
			return 0;
	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
		namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
					&unifilename, UDF_NAME_LEN);
		if (!namelen)
			return 0;
	} else
		return 0;

	return namelen;
}

#define ILLEGAL_CHAR_MARK	'_'
#define EXT_MARK		'.'
#define CRC_MARK		'#'
#define EXT_SIZE 		5

static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
				  int udfLen, uint8_t *fidName,
				  int fidNameLen)
{
	int index, newIndex = 0, needsCRC = 0;
	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	unsigned short valueCRC;
	uint8_t curr;
	const uint8_t hexChar[] = "0123456789ABCDEF";

	if (udfName[0] == '.' &&
	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
		needsCRC = 1;
		newIndex = udfLen;
		memcpy(newName, udfName, udfLen);
	} else {
		for (index = 0; index < udfLen; index++) {
			curr = udfName[index];
			if (curr == '/' || curr == 0) {
				needsCRC = 1;
				curr = ILLEGAL_CHAR_MARK;
				while (index + 1 < udfLen &&
						(udfName[index + 1] == '/' ||
						 udfName[index + 1] == 0))
					index++;
			}
			if (curr == EXT_MARK &&
					(udfLen - index - 1) <= EXT_SIZE) {
				if (udfLen == index + 1)
					hasExt = 0;
				else {
					hasExt = 1;
					extIndex = index;
					newExtIndex = newIndex;
				}
			}
			if (newIndex < 256)
				newName[newIndex++] = curr;
			else
				needsCRC = 1;
		}
	}
	if (needsCRC) {
		uint8_t ext[EXT_SIZE];
		int localExtIndex = 0;

		if (hasExt) {
			int maxFilenameLen;
			for (index = 0;
			     index < EXT_SIZE && extIndex + index + 1 < udfLen;
			     index++) {
				curr = udfName[extIndex + index + 1];

				if (curr == '/' || curr == 0) {
					needsCRC = 1;
					curr = ILLEGAL_CHAR_MARK;
					while (extIndex + index + 2 < udfLen &&
					      (index + 1 < EXT_SIZE &&
						(udfName[extIndex + index + 2] == '/' ||
						 udfName[extIndex + index + 2] == 0)))
						index++;
				}
				ext[localExtIndex++] = curr;
			}
			maxFilenameLen = 250 - localExtIndex;
			if (newIndex > maxFilenameLen)
				newIndex = maxFilenameLen;
			else
				newIndex = newExtIndex;
		} else if (newIndex > 250)
			newIndex = 250;
		newName[newIndex++] = CRC_MARK;
		valueCRC = crc_itu_t(0, fidName, fidNameLen);
		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];

		if (hasExt) {
			newName[newIndex++] = EXT_MARK;
			for (index = 0; index < localExtIndex; index++)
				newName[newIndex++] = ext[index];
		}
	}

	return newIndex;
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* unicode.c
	3	*
	4	* PURPOSE
	5	* Routines for converting between UTF-8 and OSTA Compressed Unicode.
	6	* Also handles filename mangling
	7	*
	8	* DESCRIPTION
	9	* OSTA Compressed Unicode is explained in the OSTA UDF specification.
	10	* http://www.osta.org/
	11	* UTF-8 is explained in the IETF RFC XXXX.
	12	* ftp://ftp.internic.net/rfc/rfcxxxx.txt
	13	*
1da177e4 LT	14	* COPYRIGHT
	15	* This file is distributed under the terms of the GNU General Public
	16	* License (GPL). Copies of the GPL can be obtained from:
	17	* ftp://prep.ai.mit.edu/pub/gnu/GPL
	18	* Each contributing author retains all rights to their own work.
	19	*/
	20
	21	#include "udfdecl.h"
	22
	23	#include <linux/kernel.h>
	24	#include <linux/string.h> /* for memset */
	25	#include <linux/nls.h>
f845fced	26	#include <linux/crc-itu-t.h>
1da177e4 LT	27
	28	#include "udf_sb.h"
	29
	30	static int udf_translate_to_linux(uint8_t , uint8_t , int, uint8_t *, int);
	31
28de7948	32	static int udf_char_to_ustr(struct ustr dest, const uint8_t src, int strlen)
1da177e4	33	{
cb00ea35	34	if ((!dest) \|\| (!src) \|\| (!strlen) \|\| (strlen > UDF_NAME_LEN - 2))
1da177e4	35	return 0;
28de7948	36
1da177e4 LT	37	memset(dest, 0, sizeof(struct ustr));
	38	memcpy(dest->u_name, src, strlen);
	39	dest->u_cmpID = 0x08;
	40	dest->u_len = strlen;
28de7948	41
1da177e4 LT	42	return strlen;
	43	}
	44
	45	/*
	46	* udf_build_ustr
	47	*/
28de7948	48	int udf_build_ustr(struct ustr dest, dstring ptr, int size)
1da177e4 LT	49	{
	50	int usesize;
	51
6305a0a9	52	if (!dest \|\| !ptr \|\| !size)
1da177e4	53	return -1;
6305a0a9	54	BUG_ON(size < 2);
1da177e4	55
6305a0a9 MS	56	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
6305a0a9 MS	57	usesize = min(usesize, size - 2);
cb00ea35	58	dest->u_cmpID = ptr[0];
6305a0a9 MS	59	dest->u_len = usesize;
	60	memcpy(dest->u_name, ptr + 1, usesize);
	61	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);
28de7948	62
1da177e4 LT	63	return 0;
	64	}
	65
	66	/*
	67	* udf_build_ustr_exact
	68	*/
28de7948	69	static int udf_build_ustr_exact(struct ustr dest, dstring ptr, int exactsize)
1da177e4	70	{
cb00ea35	71	if ((!dest) \|\| (!ptr) \|\| (!exactsize))
1da177e4 LT	72	return -1;
	73
	74	memset(dest, 0, sizeof(struct ustr));
cb00ea35 CG	75	dest->u_cmpID = ptr[0];
	76	dest->u_len = exactsize - 1;
	77	memcpy(dest->u_name, ptr + 1, exactsize - 1);
28de7948	78
1da177e4 LT	79	return 0;
	80	}
	81
	82	/*
	83	* udf_ocu_to_utf8
	84	*
	85	* PURPOSE
	86	* Convert OSTA Compressed Unicode to the UTF-8 equivalent.
	87	*
1da177e4 LT	88	* PRE-CONDITIONS
	89	* utf Pointer to UTF-8 output buffer.
	90	* ocu Pointer to OSTA Compressed Unicode input buffer
	91	* of size UDF_NAME_LEN bytes.
	92	* both of type "struct ustr *"
	93	*
	94	* POST-CONDITIONS
	95	* <return> Zero on success.
	96	*
	97	* HISTORY
	98	* November 12, 1997 - Andrew E. Mileski
	99	* Written, tested, and released.
	100	*/
79cfe0ff	101	int udf_CS0toUTF8(struct ustr utf_o, const struct ustr ocu_i)
1da177e4	102	{
79cfe0ff	103	const uint8_t *ocu;
1da177e4 LT	104	uint8_t cmp_id, ocu_len;
	105	int i;
	106
1da177e4	107	ocu_len = ocu_i->u_len;
cb00ea35	108	if (ocu_len == 0) {
1da177e4	109	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	110	return 0;
	111	}
	112
79cfe0ff	113	cmp_id = ocu_i->u_cmpID;
	114	if (cmp_id != 8 && cmp_id != 16) {
	115	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	116	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	117	cmp_id, ocu_i->u_name);
1da177e4 LT	118	return 0;
	119	}
	120
79cfe0ff	121	ocu = ocu_i->u_name;
79cfe0ff	122	utf_o->u_len = 0;
cb00ea35	123	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4 LT	124
1da177e4 LT	125	/* Expand OSTA compressed Unicode to Unicode */
79cfe0ff	126	uint32_t c = ocu[i++];
1da177e4 LT	127	if (cmp_id == 16)
	128	c = (c << 8) \| ocu[i++];
	129
	130	/* Compress Unicode to UTF-8 */
79cfe0ff	131	if (c < 0x80U)
28de7948	132	utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
79cfe0ff	133	else if (c < 0x800U) {
4b11111a MS	134	utf_o->u_name[utf_o->u_len++] =
	135	(uint8_t)(0xc0 \| (c >> 6));
	136	utf_o->u_name[utf_o->u_len++] =
	137	(uint8_t)(0x80 \| (c & 0x3f));
cb00ea35	138	} else {
4b11111a MS	139	utf_o->u_name[utf_o->u_len++] =
	140	(uint8_t)(0xe0 \| (c >> 12));
	141	utf_o->u_name[utf_o->u_len++] =
	142	(uint8_t)(0x80 \|
	143	((c >> 6) & 0x3f));
	144	utf_o->u_name[utf_o->u_len++] =
	145	(uint8_t)(0x80 \| (c & 0x3f));
1da177e4 LT	146	}
1da177e4 LT	147	}
cb00ea35	148	utf_o->u_cmpID = 8;
1da177e4 LT	149
	150	return utf_o->u_len;
	151	}
	152
	153	/*
	154	*
	155	* udf_utf8_to_ocu
	156	*
	157	* PURPOSE
	158	* Convert UTF-8 to the OSTA Compressed Unicode equivalent.
	159	*
	160	* DESCRIPTION
	161	* This routine is only called by udf_lookup().
	162	*
	163	* PRE-CONDITIONS
	164	* ocu Pointer to OSTA Compressed Unicode output
	165	* buffer of size UDF_NAME_LEN bytes.
	166	* utf Pointer to UTF-8 input buffer.
	167	* utf_len Length of UTF-8 input buffer in bytes.
	168	*
	169	* POST-CONDITIONS
	170	* <return> Zero on success.
	171	*
	172	* HISTORY
	173	* November 12, 1997 - Andrew E. Mileski
	174	* Written, tested, and released.
	175	*/
28de7948	176	static int udf_UTF8toCS0(dstring ocu, struct ustr utf, int length)
1da177e4 LT	177	{
	178	unsigned c, i, max_val, utf_char;
	179	int utf_cnt, u_len;
	180
	181	memset(ocu, 0, sizeof(dstring) * length);
	182	ocu[0] = 8;
	183	max_val = 0xffU;
	184
28de7948	185	try_again:
1da177e4 LT	186	u_len = 0U;
	187	utf_char = 0U;
	188	utf_cnt = 0U;
cb00ea35	189	for (i = 0U; i < utf->u_len; i++) {
28de7948	190	c = (uint8_t)utf->u_name[i];
1da177e4 LT	191
1da177e4 LT	192	/* Complete a multi-byte UTF-8 character */
cb00ea35	193	if (utf_cnt) {
1da177e4 LT	194	utf_char = (utf_char << 6) \| (c & 0x3fU);
	195	if (--utf_cnt)
	196	continue;
cb00ea35	197	} else {
1da177e4	198	/* Check for a multi-byte UTF-8 character */
cb00ea35	199	if (c & 0x80U) {
1da177e4	200	/* Start a multi-byte UTF-8 character */
cb00ea35	201	if ((c & 0xe0U) == 0xc0U) {
1da177e4 LT	202	utf_char = c & 0x1fU;
1da177e4 LT	203	utf_cnt = 1;
cb00ea35	204	} else if ((c & 0xf0U) == 0xe0U) {
1da177e4 LT	205	utf_char = c & 0x0fU;
1da177e4 LT	206	utf_cnt = 2;
cb00ea35	207	} else if ((c & 0xf8U) == 0xf0U) {
1da177e4 LT	208	utf_char = c & 0x07U;
1da177e4 LT	209	utf_cnt = 3;
cb00ea35	210	} else if ((c & 0xfcU) == 0xf8U) {
1da177e4 LT	211	utf_char = c & 0x03U;
1da177e4 LT	212	utf_cnt = 4;
cb00ea35	213	} else if ((c & 0xfeU) == 0xfcU) {
1da177e4 LT	214	utf_char = c & 0x01U;
1da177e4 LT	215	utf_cnt = 5;
28de7948	216	} else {
1da177e4	217	goto error_out;
28de7948	218	}
1da177e4	219	continue;
28de7948	220	} else {
1da177e4 LT	221	/* Single byte UTF-8 character (most common) */
1da177e4 LT	222	utf_char = c;
28de7948	223	}
1da177e4 LT	224	}
	225
	226	/* Choose no compression if necessary */
cb00ea35	227	if (utf_char > max_val) {
28de7948	228	if (max_val == 0xffU) {
1da177e4	229	max_val = 0xffffU;
28de7948	230	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	231	goto try_again;
	232	}
	233	goto error_out;
	234	}
	235
4b11111a	236	if (max_val == 0xffffU)
28de7948	237	ocu[++u_len] = (uint8_t)(utf_char >> 8);
28de7948	238	ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
1da177e4 LT	239	}
1da177e4 LT	240
cb00ea35	241	if (utf_cnt) {
28de7948	242	error_out:
1da177e4 LT	243	ocu[++u_len] = '?';
	244	printk(KERN_DEBUG "udf: bad UTF-8 character\n");
	245	}
	246
28de7948 CG	247	ocu[length - 1] = (uint8_t)u_len + 1;
28de7948 CG	248
1da177e4 LT	249	return u_len + 1;
	250	}
	251
cb00ea35	252	static int udf_CS0toNLS(struct nls_table nls, struct ustr utf_o,
34f953dd	253	const struct ustr *ocu_i)
1da177e4	254	{
34f953dd	255	const uint8_t *ocu;
1da177e4	256	uint8_t cmp_id, ocu_len;
59285c28	257	int i, len;
1da177e4	258
1da177e4 LT	259
1da177e4 LT	260	ocu_len = ocu_i->u_len;
cb00ea35	261	if (ocu_len == 0) {
1da177e4	262	memset(utf_o, 0, sizeof(struct ustr));
1da177e4 LT	263	return 0;
	264	}
	265
34f953dd	266	cmp_id = ocu_i->u_cmpID;
	267	if (cmp_id != 8 && cmp_id != 16) {
	268	memset(utf_o, 0, sizeof(struct ustr));
cb00ea35 CG	269	printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
cb00ea35 CG	270	cmp_id, ocu_i->u_name);
1da177e4 LT	271	return 0;
	272	}
	273
34f953dd	274	ocu = ocu_i->u_name;
34f953dd	275	utf_o->u_len = 0;
cb00ea35	276	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
1da177e4	277	/* Expand OSTA compressed Unicode to Unicode */
34f953dd	278	uint32_t c = ocu[i++];
1da177e4 LT	279	if (cmp_id == 16)
	280	c = (c << 8) \| ocu[i++];
	281
59285c28 JK	282	len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
	283	UDF_NAME_LEN - utf_o->u_len);
	284	/* Valid character? */
	285	if (len >= 0)
	286	utf_o->u_len += len;
	287	else
	288	utf_o->u_name[utf_o->u_len++] = '?';
1da177e4	289	}
cb00ea35	290	utf_o->u_cmpID = 8;
1da177e4 LT	291
	292	return utf_o->u_len;
	293	}
	294
28de7948	295	static int udf_NLStoCS0(struct nls_table nls, dstring ocu, struct ustr *uni,
cb00ea35	296	int length)
1da177e4	297	{
59285c28 JK	298	int len;
59285c28 JK	299	unsigned i, max_val;
1da177e4 LT	300	uint16_t uni_char;
	301	int u_len;
	302
	303	memset(ocu, 0, sizeof(dstring) * length);
	304	ocu[0] = 8;
	305	max_val = 0xffU;
	306
28de7948	307	try_again:
1da177e4	308	u_len = 0U;
cb00ea35 CG	309	for (i = 0U; i < uni->u_len; i++) {
cb00ea35 CG	310	len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
59285c28	311	if (!len)
1da177e4	312	continue;
59285c28 JK	313	/* Invalid character, deal with it */
	314	if (len < 0) {
	315	len = 1;
	316	uni_char = '?';
	317	}
1da177e4	318
cb00ea35	319	if (uni_char > max_val) {
1da177e4	320	max_val = 0xffffU;
28de7948	321	ocu[0] = (uint8_t)0x10U;
1da177e4 LT	322	goto try_again;
1da177e4 LT	323	}
cb00ea35	324
1da177e4	325	if (max_val == 0xffffU)
28de7948 CG	326	ocu[++u_len] = (uint8_t)(uni_char >> 8);
28de7948 CG	327	ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
1da177e4 LT	328	i += len - 1;
	329	}
	330
28de7948	331	ocu[length - 1] = (uint8_t)u_len + 1;
1da177e4 LT	332	return u_len + 1;
	333	}
	334
28de7948	335	int udf_get_filename(struct super_block sb, uint8_t sname, uint8_t *dname,
cb00ea35	336	int flen)
1da177e4	337	{
530f1a5e MS	338	struct ustr filename, unifilename;
530f1a5e MS	339	int len = 0;
1da177e4	340
530f1a5e MS	341	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
530f1a5e MS	342	if (!filename)
1da177e4	343	return 0;
1da177e4	344
530f1a5e MS	345	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
	346	if (!unifilename)
	347	goto out1;
	348
	349	if (udf_build_ustr_exact(unifilename, sname, flen))
	350	goto out2;
	351
cb00ea35	352	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
530f1a5e	353	if (!udf_CS0toUTF8(filename, unifilename)) {
4b11111a MS	354	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	355	sname);
530f1a5e	356	goto out2;
1da177e4	357	}
cb00ea35	358	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
530f1a5e MS	359	if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
530f1a5e MS	360	unifilename)) {
4b11111a MS	361	udf_debug("Failed in udf_get_filename: sname = %s\n",
4b11111a MS	362	sname);
530f1a5e	363	goto out2;
1da177e4	364	}
4b11111a	365	} else
530f1a5e MS	366	goto out2;
	367
	368	len = udf_translate_to_linux(dname, filename->u_name, filename->u_len,
	369	unifilename->u_name, unifilename->u_len);
	370	out2:
	371	kfree(unifilename);
	372	out1:
	373	kfree(filename);
	374	return len;
1da177e4 LT	375	}
1da177e4 LT	376
28de7948 CG	377	int udf_put_filename(struct super_block sb, const uint8_t sname,
28de7948 CG	378	uint8_t *dname, int flen)
1da177e4 LT	379	{
	380	struct ustr unifilename;
	381	int namelen;
	382
4b11111a	383	if (!udf_char_to_ustr(&unifilename, sname, flen))
1da177e4	384	return 0;
1da177e4	385
cb00ea35	386	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
28de7948	387	namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
4b11111a	388	if (!namelen)
1da177e4	389	return 0;
cb00ea35	390	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
4b11111a MS	391	namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
	392	&unifilename, UDF_NAME_LEN);
	393	if (!namelen)
1da177e4	394	return 0;
4b11111a	395	} else
1da177e4 LT	396	return 0;
	397
	398	return namelen;
	399	}
	400
	401	#define ILLEGAL_CHAR_MARK '_'
28de7948 CG	402	#define EXT_MARK '.'
	403	#define CRC_MARK '#'
	404	#define EXT_SIZE 5
1da177e4	405
4b11111a MS	406	static int udf_translate_to_linux(uint8_t newName, uint8_t udfName,
	407	int udfLen, uint8_t *fidName,
	408	int fidNameLen)
1da177e4	409	{
cb00ea35	410	int index, newIndex = 0, needsCRC = 0;
1da177e4 LT	411	int extIndex = 0, newExtIndex = 0, hasExt = 0;
	412	unsigned short valueCRC;
	413	uint8_t curr;
	414	const uint8_t hexChar[] = "0123456789ABCDEF";
	415
28de7948 CG	416	if (udfName[0] == '.' &&
28de7948 CG	417	(udfLen == 1 \|\| (udfLen == 2 && udfName[1] == '.'))) {
1da177e4 LT	418	needsCRC = 1;
	419	newIndex = udfLen;
	420	memcpy(newName, udfName, udfLen);
cb00ea35 CG	421	} else {
cb00ea35 CG	422	for (index = 0; index < udfLen; index++) {
1da177e4	423	curr = udfName[index];
cb00ea35	424	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	425	needsCRC = 1;
1da177e4 LT	426	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	427	while (index + 1 < udfLen &&
	428	(udfName[index + 1] == '/' \|\|
	429	udfName[index + 1] == 0))
1da177e4	430	index++;
4b11111a MS	431	}
	432	if (curr == EXT_MARK &&
	433	(udfLen - index - 1) <= EXT_SIZE) {
	434	if (udfLen == index + 1)
1da177e4	435	hasExt = 0;
4b11111a	436	else {
1da177e4 LT	437	hasExt = 1;
	438	extIndex = index;
	439	newExtIndex = newIndex;
	440	}
	441	}
	442	if (newIndex < 256)
	443	newName[newIndex++] = curr;
	444	else
	445	needsCRC = 1;
	446	}
	447	}
cb00ea35	448	if (needsCRC) {
1da177e4 LT	449	uint8_t ext[EXT_SIZE];
	450	int localExtIndex = 0;
	451
cb00ea35	452	if (hasExt) {
1da177e4	453	int maxFilenameLen;
4b11111a MS	454	for (index = 0;
	455	index < EXT_SIZE && extIndex + index + 1 < udfLen;
	456	index++) {
1da177e4 LT	457	curr = udfName[extIndex + index + 1];
1da177e4 LT	458
cb00ea35	459	if (curr == '/' \|\| curr == 0) {
1da177e4 LT	460	needsCRC = 1;
1da177e4 LT	461	curr = ILLEGAL_CHAR_MARK;
4b11111a MS	462	while (extIndex + index + 2 < udfLen &&
	463	(index + 1 < EXT_SIZE &&
	464	(udfName[extIndex + index + 2] == '/' \|\|
	465	udfName[extIndex + index + 2] == 0)))
1da177e4 LT	466	index++;
	467	}
	468	ext[localExtIndex++] = curr;
	469	}
	470	maxFilenameLen = 250 - localExtIndex;
	471	if (newIndex > maxFilenameLen)
	472	newIndex = maxFilenameLen;
	473	else
	474	newIndex = newExtIndex;
4b11111a	475	} else if (newIndex > 250)
1da177e4 LT	476	newIndex = 250;
1da177e4 LT	477	newName[newIndex++] = CRC_MARK;
f845fced	478	valueCRC = crc_itu_t(0, fidName, fidNameLen);
1da177e4 LT	479	newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
	480	newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
	481	newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
	482	newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
	483
cb00ea35	484	if (hasExt) {
1da177e4	485	newName[newIndex++] = EXT_MARK;
cb00ea35	486	for (index = 0; index < localExtIndex; index++)
1da177e4 LT	487	newName[newIndex++] = ext[index];
	488	}
	489	}
28de7948	490
1da177e4 LT	491	return newIndex;
1da177e4 LT	492	}